Have a sense of right and wrong as human beings

What is real intelligence? It is not a matter of executing a human order, even if it is done well. For robots to have real intelligence, they first have to have the same sense of right and wrong as human beings (moral decision), which is what Tufts University, Brown University and Rensselaer Institute want to do, and they set up a project to do research.

The project wants to simulate the moral process of human beings by creating a computational framework that allows the robot to determine autonomously what to do if it encounters a sudden-coming task. This typical scenario is in the military, such as a robot is assigned to the emergency transport mission, it is to the nearby hospital to transport drugs, but just halfway to meet a fractured wounded soldier for help, then the robot should stop to implement the rescue? How would he assess whether it would be more reasonable to stop to help the wounded, or to send the medicine to the hospital in time to save the others but leave the wounded out of the way?

So, when an unexpected event occurs, it is valuable to have the robot do a specific analysis of the measures to be taken. But because the possible scenarios cannot be defined and described with limited rules, the study will face a challenge.